Is participation in high school athletics an investment or a consumption good?: Evidence from high school and beyond

https://doi.org/10.1016/S0272-7757(00)00033-9Get rights and content

Abstract

Most high school students participate in at least one school-sponsored extracurricular activity, with varsity sports being the most popular. Conventional wisdom suggests that varsity sports participation provides participants with valuable skills. We use an instrumental variable approach to estimate whether the conventional wisdom is accurate. Using height as an instrument for participation, we find evidence that sports participation has a negative effect on the educational attainment of white male student athletes, a positive effect on the educational attainment and earnings of black male student athletes, and a positive effect on the educational attainment of white female student athletes. We find no effect of participation on the educational attainment or earnings of Hispanic males or black and Hispanic females.

Introduction

Most high school students engage in at least one of a variety of school-sponsored extracurricular activities (e.g. varsity sports, debate, chorus, band, or academic clubs). In 1992, over 80% of high school seniors participated in a school-sponsored extracurricular activity (US Department of Education, 1995). Among all high school students, regardless of gender, race, or ethnicity, the single most popular school-sponsored extracurricular activity is varsity sports. For example, 44% of male and 29% of female high school seniors participated in a varsity sport in 1992; for black, white, and Hispanic students of both genders, participation rates were 39, 36, and 31% (US Department of Education, 1995).

The conventional wisdom is that sports participation builds character, improves self-discipline and self-esteem, and teaches participants the value of teamwork (Spreitzer, 1994). If these traits enhance the future well-being of participants, then sports participation can be viewed as a human capital investment. However, since participation may take away time from other, possibly more productive, endeavors such as studying, it is possible that it may have no effect or even a negative effect on the future well-being of participants. Under this alternative scenario, sports participation can be viewed as current consumption.

In light of these two competing hypotheses of the effects of sports participation on the future well-being of participants, and in light of the high participation rates among high school students, it is somewhat surprising that economists have largely ignored the effects of varsity sports participation on the human capital accumulation of high school students. Economists have, however, studied the effect of other school-related expenditures on the human capital investment and future consumption of high school students. For example, economists have studied the relationship between school quality and earnings (Altonji & Dunn, 1995, Betts, 1995, Card & Krueger, 1992, Grogger, 1996a, Grogger, 1996b), the relationship between school inputs and academic achievement (Hanushek, 1986), and the relationship between school curriculum and educational attainment and earnings (Altonji, 1993, Altonji, 1995; Brown & Corcoran, 1996).

While these studies focus largely on factors affecting students' time in the classroom, economists have also examined the relationship between students' out-of-class activities and their human capital investment and future consumption. To date, the main focus of this body of research has been the relationship between working during school and the future educational attainment and earnings of high school students (Ehrenberg & Sherman, 1987, Griliches, 1980, Hotz, Xu, Tienda & Ahituv, 1995, Lillydahl, 1990). According to Hotz et al. (1995), however, roughly two-thirds of high school seniors worked part-time, which is less than the fraction of students who participate in any school-sponsored extracurricular activity and only slightly more than the proportion who participate in varsity sports.

In this paper, we ask if students who participate in varsity sports are investing in human capital or engaging in current consumption. If participation is an investment in human capital, as the conventional wisdom suggests, then participants will have higher levels of human capital and higher consumption in future periods. If, however, participation is current consumption without investment, then participants will appear similar to or worse off in the future than non-participants.

Previous research, conducted mainly by non-economists, has generally shown a positive association between students' varsity sports participation and their grades, school progress, and college enrollment.1 An important issue not addressed by previous researchers, however, is the potential endogeneity of varsity sports participation. If unobservable (to the researcher) factors affect both the decision to participate and the outcome analyzed, then standard statistical techniques will give inconsistent estimates of the effect of varsity sports participation on the outcome because the effect of the unobserved factor will be attributed to the observed participation decision. For example, if self-discipline is positively correlated with educational attainment, and if student athletes are more disciplined than non-athletes independent of athletic participation, then failing to account for self-discipline in a regression of educational attainment on varsity sports participation will lead one to overstate the effect of participation. If, however, student athletes have less self-discipline than non-athletes, independent of athletic participation, then one will understate the effect of participation.

We employ a two-fold estimation strategy due to the potential endogeneity of the varsity sports participation decision. First, we provide Ordinary Least Squares (OLS) estimates of the effect of participation on each of the outcomes. These estimates provide a comparison with the results of previous research and with the estimates that account for endogenous participation. To account for this endogeneity, we use an instrumental variable (IV) approach, using the students' height at age 16 as our instrument.

We realize that our IV results hinge on the validity of our instrument. To be valid, the instrument must meet two requirements. It must be correlated with the sports participation decision and uncorrelated with the outcomes. Height is correlated with sports participation because it is a valuable asset in most sports. Surely, there are exceptions such as golf, wrestling, and gymnastics to name a few, but the sports for which height is not an asset require fewer numbers of participants than say, football, basketball, volleyball, or baseball where height has clear and obvious advantages. Therefore, one would expect height to be positively and significantly correlated with sports participation. However for height at age 16 to be a valid instrument, it also must be, conditional on the other independent variables in the model, uncorrelated with the outcomes. Intuitively, this says that, conditional on the set of observed characteristics of the students, taller students are no more or no less likely to, say, graduate from high school than shorter students. We test the validity of the instrument where possible, and in the case where statistical tests are unavailable, we provide evidence which suggests height is a valid instrument.

For data, we use the nationally representative High School and Beyond data set, which provides measures of sports participation, height at age 16, educational attainment, labor market experiences, and individual- and school-level characteristics. We use the educational attainment information to construct three measures of human capital investment—dropping out of high school, college enrollment, and college graduation. We use one labor market outcome, earnings ten years after the respondent's senior year in high school, as our measure of future consumption.

In what follows, we present a statistical model of sports participation. Next, we describe the data and the sample construction, present sample statistics, and replicate some of the results of previous research. In the fourth section, we present our main results. The results are estimated separately by gender, and within gender, we provide separate estimates by race and ethnicity. We then present evidence that our instrument is valid.

Section snippets

Estimation

To test whether varsity sports participation results in higher levels of educational attainment or higher levels of future earnings, we consider the following general model,Yi=Xiβ+Piδ+εi

The dependent variable Yi represents a given outcome for the ith individual. For the three educational outcomes Yi is a binary variable, and for log annual earnings Yi is a continuous variable. The vector Xi contains a set of individual- and school-specific characteristics, Pi is an indicator variable which

Data

To examine the relationship between varsity sports participation and our education and labor market outcomes, we use the 1980 sophomore cohort of the nationally representative High School and Beyond data set (National Center for Education Statistics). Information on respondents, their families, and their high schools was collected in the 1980 base year survey and in the 1982 first follow up. Subsequent follow ups (1984, 1986, and restricted 1992) collected information on the respondents'

The estimated effect of varsity sports participation for men

The results in Table 2, Table 3 are consistent with previous work, and with the exception of the results for black females, they suggest participation is an investment which leads to greater education among participants. However, these simple comparisons may mask the effect of other observable characteristics which may be correlated with varsity sports participation and the outcomes in the table. Therefore, we turn now to our empirical estimates of the effect of participation on our educational

Conclusion

Past research, conducted primarily by non-economists, indicates that varsity sports participation provides positive future benefits to high school student athletes. However, those analyses fail to control for the endogeneity of the participation decision, and hence their estimates may be biased. Our analysis sets out to overcome this problem by using height as an instrument for the participation decision.

Our results indicate that, for white females, the general conclusion of past research is

Acknowledgements

We thank Julie DaVanzo, Jeff Grogger, Dan Hamermesh, Lynn Karoly, Robert Reville, Mary Beth Savio, Bob Schoeni, Duncan Thomas, and three anonymous referees for helpful comments. Stephen Norman provided excellent research assistance. All errors are our own.

References (27)

  • J.G. Altonji

    The demand for and return to education when education outcomes are uncertain

    Journal of Labor Economics

    (1993)
  • J.G. Altonji

    The effects of high school curriculum on education and labor market outcomes

    Journal of Human Resources

    (1995)
  • Altonji, J.G., & Dunn, T.A. (1995). The effects of school and family characteristics on the return to education. NBER...
  • Angrist, J.D. (1991). Instrumental variables estimation of average treatment effects in econometrics and epidemiology....
  • S. Averett et al.

    The economic reality of the beauty myth

    Journal of Human Resources

    (1996)
  • J.R. Betts

    Does school quality matter? Evidence from the national longitudinal survey of youth

    The Review of Economics and Statistics

    (1995)
  • J. Bound et al.

    Problems with instrumental variables estimation when the correlation between the instruments and the endogenous explanatory variable is weak

    Journal of the American Statistical Association

    (1995)
  • D.J. Brewer et al.

    Does it pay to attend an elite private college? Cross cohort evidence on the effects of college type on earnings

    Journal of Human Resources

    (1999)
  • Brown, C., & Corcoran, M. (1996). Sex-based differences in school content and the male–female wage gap. NBER working...
  • D. Card et al.

    Does school quality matter? Returns to education and the characteristics of public schools in the united states

    Journal of Political Economy

    (1992)
  • W.N. Evans et al.

    Finishing high school and starting college: do catholic schools make a difference

    Quarterly Journal of Economics

    (1995)
  • R.G. Ehrenberg et al.

    Employment while in college, academic achievement, and postcollege outcomes: a summary of results

    Journal of Human Resources

    (1987)
  • N. Fejgin

    Participation in high school competitive sports: a subversion of school mission or contribution to academic goals?

    Sociology of Sport Journal

    (1994)
  • Cited by (0)

    This paper was written while Ronan was a postdoctoral fellow at RAND. Eide was supported by a research grant from the College of Family, Home and Social Sciences at Brigham Young University. Ronan acknowledges support from the National Institute of Child Health and Human Development under grant number T32-H007329.

    View full text